Cre/lox conditional genetic approaches in the mouse are used for cell type-specific and/or temporal gene inactivation to bypass early embryonic lethal mutations or for activation of gene expression in cell type-specific and/or temporal patterns. These approaches generally rely on defined cell type-specific or ubiquitous gene regulatory sequences to direct Cre or CreER (tamoxifen-inducible) expression. Cell types or tissue regions that lack associated gene regulatory sequences or are influenced by tamoxifen are therefore difficult to genetically modify, hindering studies of gene function in these cell typs or tissues. Classical embryological manipulations have been used to modify developing tissues in the absence of specific gene regulatory sequences. For example, beads soaked in secreted proteins can be physically implanted into a precise region of a developing embryo of a specific stage to alter tissue differentiation. Similarly, tissue culture cells stably expressing a secreted protein-encoding gene can be pelleted and then transplanted into an embryonic region to alter host tissue differentiation. However, these embryological approaches are limited because they are restricted to examining genes encoding secreted proteins or molecules not encoded by genes (e.g. retinoic acid) and because the transplanted bead or cell pellet necessarily disrupts the natural architecture of the developing embryonic tissue. In addition, bead and cell pellet transplants can only be performed in a limited set of situations. We propose a genetic approach that is independent of defined cell type-specific regulatory sequences to express any gene that also maintains tissue architecture to bypass previous difficulties with standard Cre/lox approaches. This should open up novel possibilities for conditional genetic gain- and loss-of-function approaches in the mouse. Our approach should also be applicable to other model organisms. The primary objective of this proposal is to generate transgenic mice that express a light-inducible Cre system to combine with any of the current Cre-dependent gain- and loss-of-function alleles to study a wide variety of biological processes and model human diseases.